Motion-induced blurring and artifact phenomena continue to limit the potential of MR imaging in a variety of organ systems and clinical settings. These difficulties have become accentuated even as imaging systems have been improved: increases in available signal-to-noise are almost inevitably traded for improved spatial resolution or decreased acquisition time, often increasing the sensitivity of imaging examinations to the effects of motion. The importance of physiological motion is now well appreciated, though still poorly addressed with present technology. Gross motion can be equally problematic in many settings, and could indeed be considered to be the source of the greatest risks associated with MRI since it the usual motivation for the use of sedation or anesthesia in pediatric and other patient groups. Rapid MR acquisition methods show some promise for alleviating these problems, but they are associated with a number of tradeoffs; their future role for addressing the problem of motion remains hypothetical. Most of the other motion correction techniques that are currently available are helpful for reducing artifacts, but they cannot correct motion unsharpness. This project is designed to investigate a new approach for imaging moving objects with MRI. Mathematical operators are applied to the MR data prior to reconstruction. These operators adaptively correct the image data for global view-to-view tissue motion and, if desired, for bulk phase shifts caused by intraview tissue motion. They essentially transfer the frame of reference of the image coordinate system from the scanner tabletop to the moving frame. The underlying hypothesis of this approach is that although the pattern of motion is unique during any given acquisition, it often globally similar for many volume elements in a clinically relevant field of view. A prerequisite for such adaptive correction is detailed information about the motion that occurs during image acquisition. One way to obtain this information is to use specially encoded Navigator echoes, which are interleaved into the imaging sequence. Viewed together, the adaptive methods described in this proposal point to the development of an extremely robust imaging method, compatible with clinically proven spin echo technology, which essentially """"""""locks"""""""" to the region of interest during acquisition. This could obviate the need for sedation or anesthesia in some patient groups. Beyond the important objective of increasing the clinical reliability of the modality, the preliminary results suggest that the technique could provide an improved level of detail in applications that are currently limited by the effects of respiratory motion.
Showing the most recent 10 out of 24 publications
